C-11 Paint

When the mapping is complete I start to do what I call "Painting". Guess it because your just painting the livery on the car skin. I wish I had a better handle on colour. Each software displays it a bit different as well the game engine also renders it a bit different. The C-11, traditional Mercedes, is of course silver. I have a bit put it with a blue tinge, looks okay, but I will have to test it in game. I start working my way around the model, top then side, then the details. I put this preview out so you can see that the different colours I used in the panels while mapping are used then as a guide for placing the various decals and details.
As well, some things like the little running light on the front fender or the exhaust protector on the side could have been modelled in 3d. They would be interesting details, but more poly's. I have a suspision that these may not be the right details to economize. But practice will tell on that.

Wheels

Wheels, wheels, wheels ... I am not sure how the "object" engine in the game works. I cheat. A SCGT car is made of twenty different .vrl files. If anybody knows of an idiots guide to game engines and how they work, I would sure love to find out a reference. Along with the there one for exhaust, tail and head lights and a couple others. Eight of which are wheels. The cheating is that I use a set from another car and move them around to customize them.
Problem 1, the game engine simulates wheel travel as the car goes up and down hills, brakes and what not. Cute little feature that some geek worked on for months I am sure. Unfortunatley the fender profile on the real cars is such that, (where there is virtually no wheel trave),l is too low to accomadate the wheels. I have to raise the fenders to accomdate it. A pain, and makes th emodel less accurate, ya its tiny and you don't notice, I guess I will have to learn to fudge the fenders earlier in the process. The bmps are a little distorted now and I am not going to go through that process again.
Problem 2, when fixing 1, you can only see results "in game". Makes small adjustments pretty labourious.
Problem 3, choose the wrong wheels and they don't fit the width of the fender.
They are only problems because I haven't taken the time to create a full set and clone them as I need them .... take the time, get prepared, planning helps, (damn it).

Toyota's Done

Done for now, both cars in game. The low res version look spretty good in game, hardly see the transition. Should be interesting to see how many cars can run in game.
Noticed, when i started the model a year ago I modelled the Lemans style gurney flap, which is different that what was used in the championship round. Pretty tired of looking at it so, that's the way she stays.

C11 Mapping

I have discovered how I can use the Hierarchy tool to help map my textures. Man I was a dummy, this makes things much easier. I decompose the model into smaller bits. Hide the ones that I don't need at the moment, and work with a smaller set of polies. Later I will combine the bits and pieces into a couple of large objects for export. The game engine has a limit on bmp's, I beleive 500 bmps are supported. This includes the environment, (track) and all cars. Using a minimum number of bitmaps is an advantage in the more graphically intense tracks.

I am curious to see whether the split will have an impact on exporting from the subsequent tools that I have to work with. I believe I will export from AC3D, in 3ds format. It will be three objects. I will import into zMod, have to combine them into a single object. From there export the single object to a vrl format supported by the game. I have done this a bunch, but my brain always gets mixed up. Dyslexia, (actually its probably more of a brain fart). You don't want to know the hoops I go through next to get a transparent window, I will save that for another day.

I will have one more bmp, for the low res model. In total I have the Primary for main body parts, a secondary one for the other parts, (driver, wing, interior etc.), one for windows, (a requirement for the transparency process, one for tires, (which will be shared with other models) and the final one for the LR VE model. I have seen the tire maps included in the main bmp's, I may consider that as I move along.

I have kind of laid out how I map things, now I don't know if it's clear or not, but if anyone has suggestions about what may be better or easier, let me know. I am still feeling my way around and I am very open to suggestions.

Mercedes-Benz C-11

I believe the was the first year, (1990), Mercedes-Benz contested in the modern era, (since the 50's,?), without the Sauber qualifier along side the name.

The model is complete, I am just mapping the textures. I am using the hierarchy fetaure within AC3D and it has been a revelation. I think I have figured out a way to use multiple BMP's on the model without making into a mess. This will allow me two 1024x1024 .BMPS, to map the model which will keep the graphic cards from the klunge that we have been noticing when there is too many cars. I should still have enough surface area to keep the graphics resonably crisp.

Secondly polies. Ya they are trivial for real modelling .... but they just clobber the my targeted game engine. It is a circa 1999 game engine ... hey may be old but it is reliable. (sort of, kind of like me, old, and somethines reliable).

What I am doing is ensuring that I build a xxxxVE, for each model. The way the game seems to work, is it will load objects, "Just In Time", JIT, for rendering. A SCGT model has actually six different objects, A, thru E and a W.

Forget W it is a wet reflection. A thru E are displayed at different distances. When the car is far away, just a spec of colour really, it uses the E model. Now when you are in game, there may be three cars around you, all high res, 2200 polies each. But there may be anywhere from six to ten in the distance ... do the the math. The game engine chokes. Now by using a low poly model for the E version, approx 200 polies I hope to avoid this. The cost a bit of a flicker when it moves form one shape to the next. The better your low poly model the smoother your transition.

The original cars and the real "proffessionals" still do make 5 distinct models, (plus the W is 6). I am however too lazy and am trying to cut corners.

taka-Q Toyota 90C-V

Completed the taka-Q liveried Toyota 90C-V. There was just one seam issue in the transition along the side deck near the middle of the car. Has this was an older model and I was just finishing it, probably out of stubbourness more than anything I tried to fix it but ended of just leaving it. The newer models I am working on are better, its being careful not to get the "fan" configuration of polies.

I will be working on a second "DENSO" liveried car for the game. No model changes just the "painting" of the bmp.

1990 WSPC

Something I have been thinking about for a while. I love the Group C cars and 1990 in many ways had to be the best year for these beauties. There was a large contingent of "Factory" cars and some of the slickest cars yet. I am planning to include many of the key cars in the set and have modelling well underway.

The Toyota's, (warts and all), are well underway, as can be seen in preview. I plan to do the 36 and 37 cars prior to release ... somewhere.

Planned
In the "paint shed".
2 Toyota's,

Modelling complete, mapping to be completed
2 Mercedes C11's, new model, model complete started mapping.
4/6 Porsche, short tails. A variant on "Weeds" car, (with blessings). Detached tail variantion was used commonly everywhere but LeMans. The modifications are complete, just, (just...yikes), have to remap model.
2 Spice SE90C's, modelling done.

Model layout started.
2 Nissan 90CK's, just starting, going to use it as a model for a process segment of this blog.

No work as of yet.
2 Jaguar XJR-11's, again, the non-LeMans cars.
4/6 standard, fin tailed Porsches
2 Spice SE89's, much different from the 90's
2 Mazda's, they only raced LeMans but I can rewrite history in my own little world.

May do some of the le Mans variants, we will see when I get bored.

Toyota 90C-V 1990

I have had this model sitting for over a year and decieded to complete it as a break from the other stuff i was doing. I can sure tell, it really has some stinky issues, seam and the like. I have, quite possibly, tried it fix it 2 dozen times with no luck. I just don't get it. No matter everytime I try to fix it now, I screw up something else. I say to my self, self, screw it ... let er go, git it done.

Maybe I will try to fix it later, or build a new one would have taken less time. I hope that at some point I can figure out these things. Mostly it has got to do, I beleive with normal, projection and smoothing. These are commands in zMod that I use to finish up the model. Now it seems to me AC3D porbably has the same, i just haven't found it yet.

Mclaren Complete

Maclaren M1B, (1966), #4 Bruce Mclaren, #5 Chris Amon

Not the best, modern race cars are slab sided wedges, much easier to model. On a request, for a break, I deceided to take this one on. It's a 1966, Mclaren M1B. Now the problem with cars ofthis era is they were constantly evolving race to race in little details. very tough to establish a definitive model.

In game you will notice some inconsistent shading between the front and rear wheels. I don't know why this is, the BMP mapping matches so I really am unsure of reasons.

The model had a few issues, as I said, I will have to become a better modeller, smooth out the edges a bit more. Still have the ploy count way too high for the target game, have to try to understand better the way to get the,"feeling" of detail in game without actually adding them.

Must admit I did it a bit quick, just for a change from the projects where I am currently focused. The older cars were sleek and smooth with the new closed LMP rules we will see more and more of the 80's style. I love it, but canI model it ?

Tools

I am using a product called AC3D for all the base 3D modelling. It is a reasonable price, very intuitive and, at this point very well supported by an active user fourm. Nothing is free, I tried all the free ones, too much hassle. Seventy bucks, best investment I ever made, ( well for this stuff anyway).

Forum
http://www.inivis.com/forum/index.php

Product
http://www.inivis.com/

Start-up

Just thought this may be a way of controlling my projects. If I feel a reponsibility to someone out in the wide world, even if they are just in my imagination, I will finish the project. If I announce it, I will finish it. It also combines the process and the resultsm there are two different audiences for both.